Optimal Interval Type 2 Fuzzy PID Controller and its Application in Inverted Pendulum System
Ning Sheng, Xiangjian Chen, Huilin Ge, Di Li · 2019 IEEE 2nd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE) · 2019
In this paper, we propose an interval 2 fuzzy PID controller (it2f-pid) to operate the uncertain model in inverted pendulum. There are two important problems in the construction of type 2 fuzzy logic system, one is type reduction and the other is rule redundancy. The former one problem aims at calculating the center of mass of a type 2 fuzzy set. For the problem of type reduction, most of the reduction methods are computation-intensive, and an improved EKM algorithm is proposed. This algorithm improves EKM algorithm in two ways. First, change the initialization conditions at the switch point. Then, the method of searching the switch point is improved, so that it can search up and down. When it comes to the later problem, i.e., rule redundancy, the concept of normalized difference should be given first. It is presented to describe the variation of adjacent singular values to reflect the essential difference between redundant rules and important rules. Then, the number of effective singular number is determined according to the critical point of the effective singular number, and the rule is determined by TLS decomposition to construct the 2-type fuzzy model. Finally, the it2f-pid controller of the new method is compared with the it2f-pid controller of the uncertain boundary method, the simplified it2f-pid controller, and the type 1 fuzzy pid controller. Simulation and experimental results show that the performance of the controller has been significantly improved compared to the tlf-pid controller.